Authors:
Haishan Zhong
1
;
David Cho
1
;
Vladimir Pervouchine
2
and
Graham Leedham
1
Affiliations:
1
Nanyang Technological University, School of Computer Engineering, Singapore
;
2
Nanyang Technological University, School of Computer Engineering; Institute for Infocomm Research, Singapore
Keyword(s):
Speaker recognition, Feature extraction, Feature evaluation.
Related
Ontology
Subjects/Areas/Topics:
Acoustic Signal Processing
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Speech Recognition
Abstract:
Automatic speaker change point detection separates different speakers from continuous speech signal by utilising the speaker characteristics. It is often a necessary step before using a speaker recognition system. Acoustic features of the speech signal such as Mel Frequency Cepstral Coefficients (MFCC) and Linear Prediction Cepstral Coefficients (LPCC) are commonly used to represent a speaker. However, the features are affected by speech content, environment, type of recording device, etc. So far, no features have been discovered, which values depend only on the speaker. In this paper four novel feature types proposed in recent journals and conference papers for speaker verification problem, are applied to the problem of speaker change point detection. The features are also used to form a combination scheme using an SVM classifier. The results shows that the proposed scheme improves the performance of speaker changing point detection as compared to the system that uses MFCC features
only. Some of the novel features of low dimensionality give comparable speaker change point detection accuracy to the high-dimensional MFCC features.
(More)